EdiText: Controllable Coarse-to-Fine Text Editing with Diffusion Language Models

TL;DR

EdiText is a novel controllable text editing framework that combines broad and fine-grained adjustments using diffusion models, enabling precise modifications across diverse tasks like toxicity and sentiment control.

Contribution

It introduces a new controllable text editing method integrating SDEdit-based broad adjustments with a self-conditioning fine-level editing technique.

Findings

Effective control over text editing at multiple scales

Robust performance across tasks like toxicity and sentiment adjustment

Combines broad and fine editing for precise modifications

Abstract

We propose EdiText, a controllable text editing method that modifies the reference text to desired attributes at various scales. We integrate an SDEdit-based editing technique that allows for broad adjustments in the degree of text editing. Additionally, we introduce a novel fine-level editing method based on self-conditioning, which allows subtle control of reference text. While being capable of editing on its own, this fine-grained method, integrated with the SDEdit approach, enables EdiText to make precise adjustments within the desired range. EdiText demonstrates its controllability to robustly adjust reference text at a broad range of levels across various tasks, including toxicity control and sentiment control.